1 Introduction

This document shows the prediction maps for the possible infection risk of trees in the Basque Country by the following pathogens:

  • Armillaria mellea
  • Diplodia sapinea
  • Fusarium circinatum
  • Heterobasidion annosum

The following algorithms were used to create the predictions:

  • Boosted Regression Trees (BRT)
  • Generalized Additive Model (GAM)
  • Generalized Linear Model (GLM)
  • k-Nearest Neighbor (KNN)
  • Random Forests (RF)
  • Support Vector Machine (SVM)
  • Extreme Gradient Boosting (XGBOOST)

The following “algorithm-pathogen-resampling” settings did not return a performance result during performance evaluation. Due to the missing performance result for these combinations no prediction maps were created.

Pathogen Algorithm Resampling
Armillaria XGBOOST spatial/spatial
Armillaria GAM spatial/spatial

Unfortunately, XGBOOST cannot handle new factor levels in prediction data. Since we predict to the whole Basque Country using environment variables, this case occurs quite often. Variables like “soil type” and “lithology type” inherit instances which only occur in some parts of the prediction area but not within the training data. Therefore, it was not possible to create prediction maps for XGBOOST.

2 Prediction Maps

2.1 Armillaria mellea

2.1.1 GAM

## [1] NA

2.1.2 GLM

plot of chunk prediction-map-armillaria-glm

2.1.3 BRT

plot of chunk prediction-map-armillaria-brt

2.1.4 RF

plot of chunk prediction-map-armillaria-rf

2.1.5 SVM

plot of chunk prediction-map-armillaria-svm

2.1.6 KNN

plot of chunk prediction-map-armillaria-knn

2.1.7 XGBOOST

## [1] NA

2.2 Heterobasidion annosum

2.2.1 GAM

## $heterobasidion

plot of chunk prediction-map-heterobasidion-gam

2.2.2 GLM

plot of chunk prediction-map-heterobasidion-glm

2.2.3 BRT

plot of chunk prediction-map-heterobasidion-brt

2.2.4 RF

plot of chunk prediction-map-heterobasidion-rf

2.2.5 SVM

plot of chunk prediction-map-heterobasidion-svm

2.2.6 KNN

plot of chunk prediction-map-heterobasidion-knn

2.2.7 XGBOOST

## [1] NA

2.3 Diplodia sapinea

2.3.1 GAM

## $diplodia

plot of chunk prediction-map-diplodia-gam

2.3.2 GLM

plot of chunk prediction-map-diplodia-glm

2.3.3 BRT

plot of chunk prediction-map-diplodia-brt

2.3.4 RF

plot of chunk prediction-map-diplodia-rf

2.3.5 SVM

plot of chunk prediction-map-diplodia-svm

2.3.6 KNN

plot of chunk prediction-map-diplodia-knn

2.3.7 XGBOOST

## [1] NA

2.3.8 Debugging

No Temperature

plot of chunk prediction-map-diplodia-no-temp

No Precipitation

plot of chunk prediction-map-diplodia-no-precip

No hail

plot of chunk prediction-map-diplodia-no-hail

No ph

plot of chunk prediction-map-diplodia-no-ph

No soil

plot of chunk prediction-map-diplodia-no-soil

No lithology

plot of chunk prediction-map-diplodia-no-lithology

No slope

plot of chunk prediction-map-diplodia-no-slope

No pisr

plot of chunk prediction-map-diplodia-no-pisr

2.4 Fusarium circinatum

2.4.1 GAM

## $fusarium

plot of chunk prediction-map-fusarium-gam

2.4.2 GLM

plot of chunk prediction-map-fusarium-glm

2.4.3 BRT

plot of chunk prediction-map-fusarium-brt

2.4.4 RF

plot of chunk prediction-map-fusarium-rf

2.4.5 SVM

plot of chunk prediction-map-fusarium-svm

2.4.6 KNN

plot of chunk prediction-map-fusarium-knn

2.4.7 XGBOOST

## [1] NA